PDFParser - images replacer

import re
import os

class PdfParser(object):
  r = re.compile(b'^(\d+) 0 obj', re.MULTILINE)
  ro = re.compile(b'^(\d+) 0 obj\n.+?\nendobj\n', re.MULTILINE | re.DOTALL)
  ri = re.compile(b'/Subtype\s*/Image', re.MULTILINE)

  def __init__(self, filename):
    self.__filename = filename
    self.__objects = []

  def __addObject(self, data, type=0, number=-1):
    self.__objects.append( [data, type, number] )

  def __read(self):
    with open(self.__filename, "rb") as fh:
      return bytearray(fh.read())

  def parse(self):
    p, data = 0, self.__read()
    for m in re.finditer(__class__.ro, data):
      if p != m.start(): self.__addObject( data[p:m.start()] )
      p = m.end()
      t = data[m.start():m.end()]
      type = 1 if __class__.ri.search(t) else 2
      self.__addObject( t, type, int(m.group(1)) )
    if len(data) != p: self.__addObject(data[p:len(data)] )

  def getImageIndexes(self):
    idxs = []
    for i, r in enumerate(self.__objects):
      if r[1] == 1: idxs.append(i)
    return idxs

  def setData(self, idx, data):
    m = __class__.r.search(data)
    if m:
      number = self.__objects[idx][2]
      data = bytearray(str(number).encode() + data[m.end(1):])
    self.__objects[idx][0] = data

  def getData(self, idx):
    return self.__objects[idx][0]

  def write(self, filename):
    with open(filename, "wb") as fh:
      for r in self.__objects:
        fh.write(r[0])

  def log(self, filename):
    cnt = 0
    with open(filename, "wt") as fh:
      for r in self.__objects:
        fh.write( "{} {} {}\n".format(r[1], r[2], len(r[0])) )
        if r[1] == 1: cnt += 1
    return(cnt)

# ------------------------------------------------------------------------------
if __name__ == '__main__':
  fn_img = r"xxx.pdf"
  fn_txt = r"xxx + text.pdf"
  fn_rep = os.path.splitext(fn_txt)[0] + ' - replaced.pdf'
  fn_log = os.path.splitext(fn_rep)[0] + '.log'

  pdf_img = PdfParser(fn_img)
  pdf_img.parse()
  idxs_img = pdf_img.getImageIndexes()

  pdf_txt = PdfParser(fn_txt)
  pdf_txt.parse()
  idxs_txt = pdf_txt.getImageIndexes()

  if len(idxs_img) != len(idxs_txt):
    raise Exception("Images number does not match!")

  for i,j in zip(idxs_txt, idxs_img):
    pdf_txt.setData(i, pdf_img.getData(j))

  pdf_txt.write(fn_rep)
  cnt = pdf_txt.log(fn_log)
  print("images replaced:", cnt)